gmakstutis/LightRAG
1
0
Fork 0
Code Issues Pull requests Projects Releases Packages Wiki Activity Actions

Merge pull request #2171 from danielaskdd/doc-name-in-full-docs

Browse source 
Fix: Add file_path field to full_docs storage
This commit is contained in:
Daniel.y 2025-10-05 13:03:14 +08:00 • committed by GitHub
commit 1b274706d8
No known key found for this signature in database
GPG key ID: B5690EEEBB952194
2 changed files with 15 additions and 7 deletions
Show stats Download patch file Download diff file Expand all files Collapse all files

15
lightrag/kg/postgres_impl.py
View file

@ -1750,6 +1750,7 @@ class PGKVStorage(BaseKVStorage):
_data = { _data = {
"id": k, "id": k,
"content": v["content"], "content": v["content"],
"doc_name": v.get("file_path", ""), # Map file_path to doc_name
"workspace": self.workspace, "workspace": self.workspace,
} }
await self.db.execute(upsert_sql, _data) await self.db.execute(upsert_sql, _data)
@ -4588,7 +4589,8 @@ TABLES = {
SQL_TEMPLATES = { SQL_TEMPLATES = {
# SQL for KVStorage # SQL for KVStorage
"get_by_id_full_docs": """SELECT id, COALESCE(content, '') as content "get_by_id_full_docs": """SELECT id, COALESCE(content, '') as content,
COALESCE(doc_name, '') as file_path
FROM LIGHTRAG_DOC_FULL WHERE workspace=$1 AND id=$2 FROM LIGHTRAG_DOC_FULL WHERE workspace=$1 AND id=$2
""", """,
"get_by_id_text_chunks": """SELECT id, tokens, COALESCE(content, '') as content, "get_by_id_text_chunks": """SELECT id, tokens, COALESCE(content, '') as content,
@ -4603,7 +4605,8 @@ SQL_TEMPLATES = {
EXTRACT(EPOCH FROM update_time)::BIGINT as update_time EXTRACT(EPOCH FROM update_time)::BIGINT as update_time
FROM LIGHTRAG_LLM_CACHE WHERE workspace=$1 AND id=$2 FROM LIGHTRAG_LLM_CACHE WHERE workspace=$1 AND id=$2
""", """,
"get_by_ids_full_docs": """SELECT id, COALESCE(content, '') as content "get_by_ids_full_docs": """SELECT id, COALESCE(content, '') as content,
COALESCE(doc_name, '') as file_path
FROM LIGHTRAG_DOC_FULL WHERE workspace=$1 AND id IN ({ids}) FROM LIGHTRAG_DOC_FULL WHERE workspace=$1 AND id IN ({ids})
""", """,
"get_by_ids_text_chunks": """SELECT id, tokens, COALESCE(content, '') as content, "get_by_ids_text_chunks": """SELECT id, tokens, COALESCE(content, '') as content,
@ -4639,10 +4642,12 @@ SQL_TEMPLATES = {
FROM LIGHTRAG_FULL_RELATIONS WHERE workspace=$1 AND id IN ({ids}) FROM LIGHTRAG_FULL_RELATIONS WHERE workspace=$1 AND id IN ({ids})
""", """,
"filter_keys": "SELECT id FROM {table_name} WHERE workspace=$1 AND id IN ({ids})", "filter_keys": "SELECT id FROM {table_name} WHERE workspace=$1 AND id IN ({ids})",
"upsert_doc_full": """INSERT INTO LIGHTRAG_DOC_FULL (id, content, workspace) "upsert_doc_full": """INSERT INTO LIGHTRAG_DOC_FULL (id, content, doc_name, workspace)
VALUES ($1, $2, $3) VALUES ($1, $2, $3, $4)
ON CONFLICT (workspace,id) DO UPDATE ON CONFLICT (workspace,id) DO UPDATE
SET content = $2, update_time = CURRENT_TIMESTAMP SET content = $2,
doc_name = $3,
update_time = CURRENT_TIMESTAMP
""", """,
"upsert_llm_response_cache": """INSERT INTO LIGHTRAG_LLM_CACHE(workspace,id,original_prompt,return_value,chunk_id,cache_type,queryparam) "upsert_llm_response_cache": """INSERT INTO LIGHTRAG_LLM_CACHE(workspace,id,original_prompt,return_value,chunk_id,cache_type,queryparam)
VALUES ($1, $2, $3, $4, $5, $6, $7) VALUES ($1, $2, $3, $4, $5, $6, $7)

7
lightrag/lightrag.py
View file

@ -961,7 +961,7 @@ class LightRAG:
doc_key = compute_mdhash_id(full_text, prefix="doc-") doc_key = compute_mdhash_id(full_text, prefix="doc-")
else: else:
doc_key = doc_id doc_key = doc_id
new_docs = {doc_key: {"content": full_text}} new_docs = {doc_key: {"content": full_text, "file_path": file_path}}
_add_doc_keys = await self.full_docs.filter_keys({doc_key}) _add_doc_keys = await self.full_docs.filter_keys({doc_key})
new_docs = {k: v for k, v in new_docs.items() if k in _add_doc_keys} new_docs = {k: v for k, v in new_docs.items() if k in _add_doc_keys}
@ -1139,7 +1139,10 @@ class LightRAG:
# 4. Store document content in full_docs and status in doc_status # 4. Store document content in full_docs and status in doc_status
# Store full document content separately # Store full document content separately
full_docs_data = { full_docs_data = {
doc_id: {"content": contents[doc_id]["content"]} doc_id: {
"content": contents[doc_id]["content"],
"file_path": contents[doc_id]["file_path"],
}
for doc_id in new_docs.keys() for doc_id in new_docs.keys()
} }
await self.full_docs.upsert(full_docs_data) await self.full_docs.upsert(full_docs_data)